ContextLattice is the local-first control plane for long-horizon agent memory and coordination.
Progressive disclosure retrieval for long-horizon AI agents over personal knowledge bases. Three-tier model: hot cache → FTS5 index → full read. Local-first, zero dependencies.
Self-hosted continuity and collaboration substrate for autonomous agents with bounded, recoverable memory and explicit restart orientation
Repo-native operating layer for AI agents: memory, planning, proof, routing, and continuity across sessions.
Deterministic reliability-floor metrics for long-horizon agents — pass^k, reliability decay curve, meltdown onset. Zero dependencies.
Architecture notes on CLI-native autonomous coding agents, tool runtimes, orchestration, UI surfaces, and protocols.
Provide a fast, precise retrieval layer that connects AI agents with personal knowledge bases to enhance context understanding and reduce noise.
Add a description, image, and links to the long-horizon-agents topic page so that developers can more easily learn about it.
To associate your repository with the long-horizon-agents topic, visit your repo's landing page and select "manage topics."